UNPUBLISHED

UNITED STATES COURT OF APPEALS

FOR THE FOURTH CIRCUIT
JOAN H. LEWIS,
Plaintiff-Appellant,
v.
TRUSTMARK INSURANCE COMPANY
No. 98-2493
(MUTUAL); CERTIFIED SYSTEMS,
INCORPORATED; ROPER PERSONNEL
SERVICE; ROPER STAFF LEASING,
INCORPORATED,
Defendants-Appellees.
Appeal from the United States District Court
for the District of South Carolina, at Columbia.
Dennis W. Shedd, District Judge.
(CA-96-1336-3-19)
Argued: May 4, 1999
Decided: July 12, 1999
Before WILKINSON, Chief Judge, and
WILKINS and LUTTIG, Circuit Judges.
_________________________________________________________________
Affirmed by unpublished per curiam opinion.
_________________________________________________________________
COUNSEL
ARGUED: Robert Edward Hoskins, FOSTER & FOSTER, Greenville, South Carolina, for Appellant. Timothy William Bouch,
LEATH, BOUCH & CRAWFORD, L.L.P., Charleston, South Caro-

lina; Franklin Grady Shuler, Jr., TURNER, PADGET, GRAHAM &

LANEY, P.A., Columbia, South Carolina, for Appellees. ON
BRIEF: L. Joel Chastain, Terry Edward Richardson, Jr., NESS,
MOTLEY, LOADHOLT, RICHARDSON & POOLE, Barnwell,
South Carolina, for Appellant. Anita M. Alessandra, William O. Ashcraft, ASHCRAFT LAW FIRM, Dallas, Texas, for Appellees Certified Systems, Roper Personnel, and Roper Leasing.
_________________________________________________________________
Unpublished opinions are not binding precedent in this circuit. See
Local Rule 36(c).
_________________________________________________________________
OPINION
PER CURIAM:
Joan Lewis was diagnosed with breast cancer and sought coverage
under her health insurance plan for a course of treatment recommended by her physician. After consulting three outside physicians
with expertise in the treatment of cancer, the administrator of the plan
determined that the treatment was experimental and denied coverage.
Lewis brought an ERISA suit challenging that denial, and the district
court granted summary judgment to the defendants. Because we hold
that the plan administrator did not abuse its discretion by finding the
treatment experimental as that term is defined by the plan, we affirm.
I.
Joan Lewis is the owner and operator of a retail gift shop named
Santee Shoppe, Ltd. in Columbia, South Carolina. Lewis claims that
in 1994, she was approached by an agent of Roper Personnel Service
and Roper Staff Leasing, Inc. (collectively Roper). The salesman proposed a contract whereby all Santee Shop employees would become
employees of another company, Certified Systems, Inc. (CSI). CSI
would handle all administrative responsibilities for the employees,
including payroll tasks, human resource services, and employee insurance and benefits. The employees would then be leased back to San2

tee Shoppe. Lewis, on behalf of Santee Shoppe, entered into the

contract with CSI. As a result, Lewis herself became an employee of
CSI.
As part of its benefit package, CSI provides a Trustmark Insurance
Company health insurance policy. Lewis became a member of the
Trustmark plan effective February 1, 1995.
Prior to becoming a member of the Trustmark plan, Lewis was
diagnosed with breast cancer. In 1989 she underwent a mastectomy
and other treatment and the cancer went into remission. After she
joined the Trustmark plan, however, the disease returned. In September 1995 Lewis was diagnosed with Stage IV metastatic breast cancer. Her treating physician, Dr. Henslee-Downey, recommended she
undergo high dose chemotherapy supported by peripheral stem cell
rescue (HDC/PSCR). Lewis then submitted to Trustmark a pretreatment authorization request for HDC/PSCR treatments.
Trustmark's Medical Director referred Lewis' request to an outside
medical review board. The board consisted of three oncologists
selected by the Medical Ombudsman Program, an independent company. Based on their opinions the plan administrator determined that
Lewis' treatment was, under the terms of the plan, medically unnecessary, investigational, and experimental. Trustmark denied coverage
for HDC/PSCR in February 1996.
Lewis appealed Trustmark's denial of benefits and submitted medical literature and affidavits supporting her position. 29 U.S.C.
1133. Trustmark reaffirmed its decision.
Lewis then filed suit against Trustmark, CSI, and Roper. Among
other things, she alleged that Trustmark improperly denied coverage.
29 U.S.C. 1132(a)(1)(B). She asserted that HDC/PSCR was medically necessary, not experimental, and thus covered by the plan.
Lewis also claimed that under South Carolina law she was fraudulently induced into joining the insurance plan by representations that
she would have full and comprehensive health benefits.
The district court granted summary judgment to the defendants. It
found that the plan administrator did not abuse its discretion by find3

ing that Lewis' treatment was not covered by the Trustmark plan. The
court also held that her fraudulent inducement claim was preempted
by ERISA. Lewis appeals.
II.
The Trustmark plan delegates to the plan administrator "full, exclusive and discretionary authority to determine all questions arising in
connection with the group contract including its interpretation." Trustmark Plan at 58. As such, we review the administrator's decision only
for abuse of discretion. Firestone Tire & Rubber Co. v. Bruch, 489
U.S. 101, 111 (1989). A reviewing court will not disturb an administrator's reasonable interpretation and application of a plan provision.
Fox v. Fox, 167 F.3d 880, 883 (4th Cir. 1999). Where, as here, the
administrator has a financial interest in the outcome of its determination, "deference will be lessened to the degree necessary to neutralize
any untoward influence resulting from the conflict" of interest. Doe
v. Group Hospitalization & Med. Servs., 3 F.3d 80, 87 (4th Cir.
1993).
A.
Lewis initially argues that the plan's incontestability clause
requires that the plan pay for her treatment. That clause states:
TIME LIMIT ON CERTAIN DEFENSES -- FOR ALL
BENEFITS OTHER THAN LIFE.
After coverage has been in force during a person's lifetime
for one year from his effective date only fraudulent misstatements in his application or enrollment form may be
used to void his coverage or to deny any claim made by him
for loss incurred starting after such one year period.
Trustmark Plan at 57. Lewis contends that because her treatment
began on February 12, 1996 -- more than one year after coverage had
been in force -- this clause bars the plan administrator from denying
a claim for benefits for any reason other than fraudulent misstatements. Thus, under the plan Trustmark has no discretion to deny her
claim for benefits.
4

We disagree. An incontestability clause may defeat an insurer's

attempt to declare a person ineligible for coverage under the plan.
Rapak v. Companion Life Ins. Co., 990 F.2d 801, 803-04 (4th Cir.
1993). But such a clause does not operate to define plan benefits or
defeat exclusions from coverage. This is so even where the incontestability clause is written in broad and expansive terms. "An incontestability clause prevents an insurer from contesting the validity of an
insurance contract. However, such a clause certainly does not prevent
the insurer from invoking the plain terms of an ERISA plan." White
v. Provident Life & Accident Ins. Co., 114 F.3d 26, 28-29 (4th Cir.
1997).
Thus, we hold that the incontestability clause at issue here does not
force the plan administrator to pay any claim without limitation. This
interpretation gives effect to the plain intent of the parties and gives
meaning to the contract as a whole. Courts "should not torture the
meaning of policy language in order to extend or defeat coverage that
was never intended by the parties." Gambrell v. Travelers Ins. Co.,
310 S.E.2d 814, 816 (S.C. 1983). To interpret the clause otherwise
would substitute an unsupportable reading of the clause for a reasonable reading of the contract.
B.
Lewis also argues that the plan administrator abused its discretion
by finding that the treatment was experimental. She asserts that
HDC/PSCR was a generally accepted treatment for Stage IV breast
cancer. She points to a South African study indicating HDC/PSCR's
superiority to conventional standard dose treatment. And Lewis notes
that her own treating physician, Dr. Henslee-Downey, stated "unequivocally that HDC/PSCR is a `generally accepted' alternative for
the treatment of Stage IV breast cancer." Finally, Lewis cites our
decision in Wilson v. Office of Civilian Health & Med. Programs of
the Uniformed Servs., 65 F.3d 361 (4th Cir. 1995). In Wilson, this
court noted that "HDC/PSCR is gaining widespread acceptance
within the medical community" and the evidence suggests "a broad
consensus of physicians recognize[s] [its] efficacy." Id. at 366 (internal quotation marks omitted). Thus, Lewis argues that HDC/PSCR
5

was not experimental and should have been covered by the Trustmark
plan.1
A court's role is to examine the ruling of a plan administrator and
decide whether that determination was a reasonable interpretation of
the written terms of the plan. Here, we cannot conclude that the Trustmark administrator's interpretation was unreasonable. It may be, as
Lewis contends, that many doctors did not consider HDC/PSCR to be
experimental, as that term is generally used in the medical profession.
But the written Trustmark benefit plan gives the term experimental a
specific definition, which governs this case. The plan explicitly states
"No benefits are paid for . . . Experimental treatment, including treatment with new drugs or technological medical devices which are
Experimental in nature." Trustmark Plan at 12. The policy then
defines experimental:
A drug, device or medical treatment or procedure is Experimental . . . if Reliable Evidence shows that the drug, device
or medical treatment or procedure is the subject of ongoing
Phase I, II, or III clinical trials or under study to determine
its maximum tolerated dose, its toxicity, its safety, its efficacy, or its efficacy as compared with the standard means of
treatment or diagnosis . . . .
Trustmark Plan at 3 (emphasis added).
_________________________________________________________________
1 In addition, Lewis argues that the consistent enforcement of the plan's
definition of experimental would deny coverage to many advanced medical procedures and that Trustmark cannot arbitrarily choose which procedures to cover and which not to cover. While this may be true, Lewis
adduced no evidence demonstrating that Trustmark arbitrarily does cover
some experimental treatments but not others.
Lewis also contends that Trustmark's own internal procedures dictate
payment for HDC/PSCR for the treatment of stage IV breast cancer for
beneficiaries in 10 states, but not for those in South Carolina. Trustmark,
however, determined that state law in those states required this coverage,
and the plan notes that the experimental treatment exclusion was "Subject to State Approval." Trustmark Plan at 12. Lewis has pointed to no
evidence indicating that Trustmark has abused its discretion in attempting to comply with state law.
6

In order to evaluate whether HDC/PSCR was experimental under

this definition, Trustmark turned to the Medical Ombudsman Program, an independent company. The Medical Ombudsman Program
in turn referred the matter to a review board of three oncologists.
Among the questions asked of the oncologists was"Are the drugs,
therapies, or treatments proposed currently the subject of ongoing
Phase I, Phase II, or Phase III clinical trials or otherwise under study
to determine the maximum tolerated dose, toxicity, safety, efficacy,
or efficacy as compared with standard[ ] treatments?"
In response to this and other questions, all three doctors indicated
that HDC/PSCR was subject to ongoing Phase I, II or III clinical trials. Dr. Samuel M. Silver, Director of the Adult Bone Marrow Transplant Program at Comprehensive Cancer Center of the University of
Michigan, stated that although he personally did not consider the
treatment experimental, "[s]ince high-dose chemotherapy needs to
undergo phase 3 studies in high priority National Cancer Institute protocols, the use of this therapy is investigational in that data is still
being collected." When asked specifically whether the treatment was
still in clinical trials, he stated, "YES. As mentioned, there are a number of phase 3 studies still ongoing and are high-priority National
Cancer Institute approved comparing transplant to lower dose therapy. The fact that this patient is receiving Gamma Interferon and
Cyclosporin makes this a phase 2 study which looks at the efficacy
of these adjunctive immunologic drugs."
To the same question Dr. Robert K. Stuart of The Medical University of South Carolina responded, "Yes. The treatment, or at least
some aspects of it [the use and dosages of certain drugs] are apparently in Phase I or Phase II clinical trials at the University of Colorado and/or the University of South Carolina."
Similarly, Dr. Robert Dreicer, Associate Professor of Medicine and
Urology at the University of Iowa, responded, "Yes with caveats. As
a larger issue there are several phase III trials including the NCI
[National Cancer Institute] sponsored intergroup study ongoing comparing the outcomes of patients with metastatic disease treated with
standard therapy versus high dose therapy with rescue. As is indicated
in Dr. Henslee-Downey's note of 10/30/95 the high dose combination
7

proposed here was developed at the University of Colorado and may

still be in phase II trials at that institution."
In sum, all three of the medical experts consulted by the Medical
Ombudsman Program concluded that the HDC/PSCR treatment prescribed for Lewis was still in clinical trials. Their opinions were reinforced by a report by the National Cancer Institute (NCI) entitled
Current Clinical Trials: Oncology, published in July/August 1995.
That publication reported an ongoing study of HDC/PSCR:
NCI HIGH PRIORITY CLINICAL TRIAL -- Phase III
Randomized Comparison of Conventional CMF (CTX/
MTX/5-FU) Maintenance vs High-Dose Chemotherapy with
CTX/TSPA/CBDCA plus Autologous Bone Marrow and
Peripheral Stem Cell Rescue in Women with Metastatic
Breast Cancer Responding to Conventional Induction Chemotherapy.
The existence of a clinical trial comparing HDC/PSCR with conventional therapy means that HDC/PSCR was still "the subject of ongoing Phase I, II, or III clinical trials." Therefore, HDC/PSCR was
experimental and thus excluded from coverage under the plain terms
of the Trustmark plan. By consulting independent experts and comparing their responses to the plain terms of the plan, the plan administrator acted reasonably.2 The district court in turn properly declined
to disturb the administrator's decision.
III.
In addition, Lewis appeals the dismissal of her fraudulent inducement claim. As noted, the district court held that this claim was preempted by ERISA. We need not address preemption because the
record indicates that this claim must fail as a matter of state law.
Lewis alleged in her complaint that "CSI and Roper approached
Plaintiff in 1994 and represented to her that if she became an
employee of CSI, she would be provided full and comprehensive cov_________________________________________________________________
2 By holding that the plan administrator reasonably found that
HDC/PSCR was experimental, we need not address the administrator's
alternative grounds for denying coverage under the plan.
8

erage for medical treatment for her breast cancer." CSI and Roper
contend that this claim is factually flawed and that no one ever made
a specific representation that HDC/PSCR treatment would be covered.
Summary judgment on the fraudulent inducement claim is certainly
appropriate. Under Federal Rule of Civil Procedure 9(b), "the circumstances constituting fraud or mistake shall be stated with particularity." See also S.C. R. Civ. P. 9(b). Once past the pleading stage, a
party must then prove the elements of fraud by clear and convincing
evidence. Kerr v. State Farm Fire and Cas. Co. , 731 F.2d 227, 22829 (4th Cir. 1984). Lewis has fallen far short of meeting this burden
of proof. To establish fraud, there must be a false representation that
is "predicated upon misstatements of fact rather than upon an expression of opinion, an expression of intention or an expression of confidence that a bargain will be satisfactory." Bishop Logging Co. v. John
Deere Indus. Equip. Co., 455 S.E.2d 183, 187 (S.C. Ct. App. 1995).
Statements of a vague and general character cannot be read to make
representations of a specific character. See, e.g., Miller v. Premier
Corp., 608 F.2d 973, 981 (4th Cir. 1979) ("[A]n unspecific and false
statement of opinion such as occurs in puffing generally cannot constitute fraud."). Otherwise, claims such as Lewis' would render
actionable a host of general positive statements about product quality.
The terms "full" and "comprehensive" are vague, general terms that
cannot be construed to make specific representations about the coverage of a particular form of experimental treatment.
Although Lewis now asserts on appeal that CSI and Roper made
representations concerning coverage for HDC/PSCR, her complaint is
at odds with her brief. She neither alleges in her complaint nor points
to any evidence in the record demonstrating that any such representation was made. Instead, she claims to have relied on the generalized
puffing of an employee lease-back sales agent about the overall quality of the health insurance coverage. A plaintiff"cannot simply cry
fraud and thereby escape summary judgment." Strum v. Exxon Co., 15
F.3d 327, 331 (4th Cir. 1994). Because Lewis has failed to point to
evidence demonstrating either a false statement of fact or that she reasonably relied on any such statement, we affirm the dismissal of her
fraudulent inducement claim.
9

IV.
For the foregoing reasons, we affirm the judgment of the district
court.
AFFIRMED
10